Complex Texture Features for Glaucoma Diagnosis Using Support Vector Machine
In this paper, Gray Level Co-occurrence Matrix (GLCM) features are effectively utilized for glaucoma diagnosis. Early diagnosis of glaucoma is important to protect vision loss. The proposed system uses four GLCM features such as Contrast, Correlation, Energy, and Homogeneity for the diagnosis. In order to effectively use these features for glaucoma detection they are extracted using the optical density transformed fundus image along with the original features. The classification of fundus image into normal or abnormal is obtained by the Support Vector Machine (SVM) classifier. An internal database of 200 images is utilized for the performance analysis. The results show that the proposed approach helps the ophthalmologists to make their decision very accurately. The proposed system provides 95% classification accuracy.
Glaucoma, Colour Fundus Image, GLCM, SVM Classifier, Optical Density Image.
Copyright (c)2015   Srinivasan C, Suneel Dubey, Ganeshbabu TR
This work is licensed under a Creative Commons Attribution 4.0 International License.